Date: Mon, 28 Oct 1996 17:09:51 -0800 (PST)
From: Chris Newman <Chris.Newman@innosoft.com>
Subject: re: X-Sender problem
In-reply-to: <19961029005227.18391.qmail@koobera.math.uic.edu>
To: Directed Revision/Update of Message Standards <drums@cs.utk.edu>
Message-id: <Pine.SOL.3.95.961028170715.11550B-100000@eleanor.innosoft.com>
MIME-version: 1.0
Content-type: TEXT/PLAIN; charset=US-ASCII
Content-transfer-encoding: 7BIT

On Tue, 29 Oct 1996, D. J. Bernstein wrote:
> I don't get it. Why is anybody using Sender automatically?

Because it looks like an email address.  And because they can.

> The RFC 822 Sender field is (aside from the nonsense about it being
> authenticated) a reasonable concept, comparable to the ``/jtw'' used on
> physical letters to indicate that JTW did the typing and mailing.

It's more than that since it is a deliverable address, but I agree.

> Putting Sender in every message, as Netscape and certain other programs
> do, seems to be a corruption of the original semantics.

Agreed.

> It's also pointless. If you want to record extra information so that
> people can trace problems, add a Received field---that's what Received
> was designed for.

I think the MUA/MTA separation is important.  Received headers are for
MTAs.  Originator-Info is an MUA generated field.

Thank you for your comments.